CONTENTS
-------------------
-------------------

* FIA_API_Final.py
* FIA_GIS_Final.py
* FIA_PSM_Final.py

-------------------
-------------------


* FIA_API_Final.py
-------------------
This file uses FIADB Application Programming Interface to query tree-, condition-, and plot-level tables. The
script is currently setup to query inventory years 2000-2019, but that can be adjusted. Official 
documentation from the US. Forest Service should be reviewed for definitions of the queried columns.
Non-timberland plots are filtered out of the dataset and timberland conditions are expanded to represent 100%
of the plot area and subsequently expanded to a per hectare basis. Above- and below-ground carbon in live and
standing dead trees is derived from tree-level measurements. Carbon in down and dead trees is derived at the
condition level. Soil and understory carbon is collected at the condition level and aggregated to the plot
level. Soil and litter carbon is supplemented from external data sets (see paper), and excluded from this script.
Authored by Houston Sudekum.

Response variables

down_dead_carbon : carbon in down and dead trees
under_crabon : carbon in the understory
livetree_carbon : carbon in live-trees
deadtree_carbon : carbon in standing dead trees

Explanatory variables

stat_fed : state/federal ownership on the largest timberland condition
art_regen : evidence of artificial regeneration on any timberland condition
aspect : aspect of the largest timberland condition on the plot
fortypcd : forest type code of the largest timberland condition of the plot
ins_dis : evidence of insect or disease damage on any timberland condition
fire_wth : evidence of fire or weather damage on any timberland condition
ecosubcd : ecological subsection of the largest timberland condition
H : shannon diversity index


* FIA_GIS_Final.py
-------------------
This script provides a function used to calculate geodesic distances. The formula was borrowed from online forums
and validated independently. This formula was used to calculate distances between FIA plots and forest product
industries.


* FIA_PSM_Final.py
-------------------
The PSM script includes a basic mock-up of probit regression and a function for PSM matching. The PSM function
was adapted from this jupyter notebook - https://nbviewer.org/github/kellieotto/StatMoments/blob/master/PSM.ipynb.
The PSM functions allows you to match with/without sampling, match 1:1/1:many, keep control with/without a match, and 
resample from the non-control group. The function was built to accept a Pandas data frame, but could be easily
altered to accept other data types. Authored by Houston Sudekum.
